wayback_machine_downloader_straw 2.3.6 → 2.3.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 04ac6f9f045b4f92a7481ad8544f2f9138454b9eabdcf6f47b28195c1dd1cdaf
-  data.tar.gz: '09a16685d1299afb338d86495d1c58825482a6785e7e1a596bb02eb2da1fc7f1'
+  metadata.gz: b739c4ecda1e325f9d5a33872fa71a8a5103f1770cc18c7e1b46516c96c8fef6
+  data.tar.gz: 991cf1f67783f35a8da233e6d9e82edc4d933ef0229d5ecffbe8963c5d049c98
 SHA512:
-  metadata.gz: fd157e047c8631ff5cdfd4ca540840a7d49196131dc4de9f9725c3989164151e4c05dda0dae0dc884bfb9bbb51483f061378ef7a1e737b36d1d11882719bcf60
-  data.tar.gz: e9b814bbbed6caef69972b9e94891f7af9be61674cf50bdd3bb1bf4a60c3622156e93b07de8a3761dba87a852bd67aa10439481c4ca72bffe564019f04451ed5
+  metadata.gz: f9b71d59d4c5c5bdb82f58fceacd848242a34b12d15abf93c101e4d61ab8fcab46e60011b80f966b0851474160af153c92ab46db5ed2c2e80b0fec3afdc53f8c
+  data.tar.gz: 88f39d47bb8405f682ddca4236bd2e3ce93ffbfd426c2430532b904c98e7cb1593406271fa4453847ab95615adbffc36049072bd7c8b45b171e2cecb77bb41ab

data/lib/wayback_machine_downloader.rb

@@ -113,7 +113,7 @@ class WaybackMachineDownloader
 
   include ArchiveAPI
 
-  VERSION = "2.3.6"
+  VERSION = "2.3.7"
   DEFAULT_TIMEOUT = 30
   MAX_RETRIES = 3
   RETRY_DELAY = 2
@@ -477,8 +477,8 @@ class WaybackMachineDownloader
         begin
           @connection_pool.with_connection do |connection|
             result_message = download_file(file_remote_info, connection)
-            # for now, assume success if no exception and message doesn't indicate error/skip
-            if result_message && !result_message.downcase.include?('error') && !result_message.downcase.include?('failed') && !result_message.downcase.include?('skipping') && !result_message.downcase.include?('already exists')
+            # assume download success if the result message contains ' -> '
+            if result_message && result_message.include?(' -> ')
                download_success = true
             end
             @download_mutex.synchronize do
@@ -659,11 +659,21 @@ class WaybackMachineDownloader
 
     begin
       structure_dir_path dir_path
-      download_with_retry(file_path, file_url, file_timestamp, http)
-      if @rewrite && File.extname(file_path) =~ /\.(html?|css|js)$/i
-        rewrite_urls_to_relative(file_path)
+      status = download_with_retry(file_path, file_url, file_timestamp, http)
+
+      case status
+      when :saved
+        if @rewrite && File.extname(file_path) =~ /\.(html?|css|js)$/i
+          rewrite_urls_to_relative(file_path)
+        end
+        "#{file_url} -> #{file_path} (#{@processed_file_count + 1}/#{@total_to_download})"
+      when :skipped_not_found
+        "Skipped (not found): #{file_url} (#{@processed_file_count + 1}/#{@total_to_download})"
+      else
+        # ideally, this case should not be reached if download_with_retry behaves as expected.
+        @logger.warn("Unknown status from download_with_retry for #{file_url}: #{status}")
+        "Unknown status for #{file_url}: #{status} (#{@processed_file_count + 1}/#{@total_to_download})"
       end
-      "#{file_url} -> #{file_path} (#{@processed_file_count + 1}/#{@total_to_download})"
     rescue StandardError => e
       msg = "Failed: #{file_url} # #{e} (#{@processed_file_count + 1}/#{@total_to_download})"
       if File.exist?(file_path) and File.size(file_path) == 0
@@ -714,8 +724,7 @@ class WaybackMachineDownloader
 
       response = connection.request(request)
 
-      case response
-      when Net::HTTPSuccess
+      save_response_body = lambda do
         File.open(file_path, "wb") do |file|
           body = response.body
           if response['content-encoding'] == 'gzip' && body && !body.empty?
@@ -725,26 +734,48 @@ class WaybackMachineDownloader
               gz.close
               file.write(decompressed_body)
             rescue Zlib::GzipFile::Error => e
-              @logger.warn("Failure decompressing gzip file #{file_url}: #{e.message}")
+              @logger.warn("Failure decompressing gzip file #{file_url}: #{e.message}. Writing raw body.")
               file.write(body)
             end
           else
             file.write(body) if body
           end
         end
-      when Net::HTTPRedirection
-        raise "Too many redirects for #{file_url}" if redirect_count >= 2
-        location = response['location']
-        @logger.warn("Redirect found for #{file_url} -> #{location}")
-        return download_with_retry(file_path, location, file_timestamp, connection, redirect_count + 1)
-      when Net::HTTPTooManyRequests
-        sleep(RATE_LIMIT * 2)
-        raise "Rate limited, retrying..."
-      when Net::HTTPNotFound
-        @logger.warn("File not found, skipping: #{file_url}")
-        return
-      else
-        raise "HTTP Error: #{response.code} #{response.message}"
+      end
+
+      if @all
+        case response
+        when Net::HTTPSuccess, Net::HTTPRedirection, Net::HTTPClientError, Net::HTTPServerError
+          save_response_body.call
+          if response.is_a?(Net::HTTPRedirection)
+            @logger.info("Saved redirect page for #{file_url} (status #{response.code}).")
+          elsif response.is_a?(Net::HTTPClientError) || response.is_a?(Net::HTTPServerError)
+            @logger.info("Saved error page for #{file_url} (status #{response.code}).")
+          end
+          return :saved
+        else
+          # for any other response type when --all is true, treat as an error to be retried or failed
+          raise "Unhandled HTTP response: #{response.code} #{response.message}"
+        end
+      else # not @all (our default behavior)
+        case response
+        when Net::HTTPSuccess
+          save_response_body.call
+          return :saved
+        when Net::HTTPRedirection
+          raise "Too many redirects for #{file_url}" if redirect_count >= 2
+          location = response['location']
+          @logger.warn("Redirect found for #{file_url} -> #{location}")
+          return download_with_retry(file_path, location, file_timestamp, connection, redirect_count + 1)
+        when Net::HTTPTooManyRequests
+          sleep(RATE_LIMIT * 2)
+          raise "Rate limited, retrying..."
+        when Net::HTTPNotFound
+          @logger.warn("File not found, skipping: #{file_url}")
+          return :skipped_not_found
+        else
+          raise "HTTP Error: #{response.code} #{response.message}"
+        end
       end
 
     rescue StandardError => e

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: wayback_machine_downloader_straw
 version: !ruby/object:Gem::Version
-  version: 2.3.6
+  version: 2.3.7
 platform: ruby
 authors:
 - strawberrymaster
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2025-05-18 00:00:00.000000000 Z
+date: 2025-05-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: concurrent-ruby